Serodiagnosis efficacy and immunogenicity of the fusion protein of Mycobacterium tuberculosis composed of the 10-kilodalton culture filtrate protein, ESAT-6, and the extracellular domain fragment of PPE68

Abstract

In order to identify immunodominant antigens of Mycobacterium tuberculosis that may be used in the serodiagnosis of active tuberculosis (TB), we designed an M. tuberculosis fusion protein consisting of CFP-10 (10-kDa culture filtrate protein), ESAT-6 (6-kDa early secreted antigenic target), and the extracellular domain fragment of PPE68 (PPE68'). Then, the coding sequences of the three proteins were inserted into a prokaryotic expression vector, pET-32a(+). To enhance the immunological response, the proteins were linked together. The fusion proteins with a 6 × His tag were successfully overexpressed in Escherichia coli BL21 and purified. The purified proteins were applied for detection of the total IgG titer by using an enzyme-linked immunosorbent assay (ELISA) with human sera from well-characterized TB cases and the control cases, and results were compared to those with purified protein derivative tuberculin (PPD). The ELISA results showed that among 140 cases of confirmed active TB and 70 control cases, CFP-10-ESAT-6-PPE68' had a sensitivity of 73.3% and specificity of 94.3%, compared to a sensitivity of 66.7% and specificity of 74.3% for PPD and a sensitivity of 65% and specificity of 91.4% for CFP-10-ESAT-6. In addition, the fusion protein CFP-10-ESAT-6-PPE68' stimulated a higher level of antigen-specific gamma interferon (IFN-γ) release for active-TB patients than PPD and CFP-10-ESAT-6. After immunization of C57BL/6 mice, the findings indicated that the total IgG titers and the concentrations of IFN-γ in mice immunized by CFP-10-ESAT-6-PPE68' were high and induced strong, long-term humoral immunity compared to results with PPD and CFP-10-ESAT-6. Thus, our study indicates that the fusion protein CFP-10-ESAT-6-PPE68' may be useful as an immunodominant antigen for the serodiagnosis of active TB.

Fig 1

Epitope analysis of the PPE68 protein, using the DNAstar software program (A) or the HMMMTOP method (B). The extracellular domain of PPE68 was base pairs 1 to 348 bp (circled), analyzed by both DNAStar software and the HMMMTOP method.

Fig 2

Agarose gel electrophoresis for PCR products. Lane M, DNA marker (DL2000); lane 1, PCR product of cfp10 (303 bp); lane 2, PCR product of esat6 (288 bp); lane 3, PCR product of ppe68′ (348 bp); lane 4, PCR product of cfp10-esat6 (693 bp); lane 5, PCR product of cfp10-esat6-ppe68′ (1,041 bp).

Fig 3

SDS-PAGE (A) or Western blot (B) analysis of the purified proteins CFP-10–ESAT-6 and CFP-10–ESAT-6–PPE68′. (A) Lane M, protein marker; lane 1, CFP-10–ESAT-6 eluted by 300 mM imidazole elution buffer (37 kDa); lane 2, CFP-10–ESAT-6 eluted by 400 mM imidazole elution buffer; lane 3, CFP-10–ESAT-6 eluted by 500 mM imidazole elution buffer; lane 4, CFP-10–ESAT-6–PPE68′ eluted by 300 mM imidazole elution buffer (49 kDa). (B) Lane M, protein marker; lane 1, the purified protein CFP-10–ESAT-6–PPE68′ (49 kDa); lane 2, the purified protein CFP-10–ESAT-6 (37 kDa); 30 mM imidazole in binding and washing buffers and 300 mM imidazole in elution buffer were the optimal choice in our study. The fusion proteins CFP-10–ESAT-6 and CFP-10–ESAT-6–PPE68′ were recognized by the rabbit serum polyclonal antibody of M. tuberculosis.

Fig 4

Area under the ROC curve for three antigens with sera from TB patients. The area under the ROC curve was highest for CFP-10–ESAT-6–PPE68′ (0.862), followed by CFP-10–ESAT-6 (0.829) and PPD (0.712). According to the ROC analysis, there was a statistically significant difference between PPD detection and CFP-10–ESAT-6–PPE68′ detection (P = 0.005) in a pairwise comparison of ROC curves. Likewise, there was a statistical difference between PPD detection and CFP-10–ESAT-6 detection (P = 0.029). Although the difference between CFP-10–ESAT-6 detection and CFP-10–ESAT-6–PPE68′ detection was not statistically significant (P = 0.491), the sensitivity and specificity of CFP-10–ESAT-6–PPE68′ were better than those of CFP-10–ESAT-6.

Fig 5

Immunogenicities of PPD, CFP-10–ESAT-6, and CFP-10–ESAT-6–PPE68′ in TB patients. The levels of IFN-γ in all samples without antigen stimuli were below 20 pg/ml (14.1 ± 3.7 pg/ml) but increased significantly to 1,272.8 ± 140.1 pg/ml after CFP-10–ESAT-6–PPE68′ stimulation, compared to 263.4 ± 34.1 pg/ml after PPD stimulation and 941.3 ± 74.5 pg/ml after CFP-10–ESAT-6 stimulation. The IFN-γ level in samples activated by CFP-10–ESAT-6–PPE68′ was significantly higher than that in samples activated by PPD and CFP-10–ESAT-6 (P < 0.05).

Fig 6

Titers of IgG of mice immunized with PPD, CFP-10–ESAT-6, and CFP-10–ESAT-6–PPE68′. The total IgG titers of CFP-10–ESAT-6–PPE68′ remained high at >5,000 between 3 and 5 weeks after enhanced immunization; those for the other groups were no more than 1,000 in the 7 weeks after enhanced immunization. Means and standard errors are shown.

Fig 7

Level of antigen-specific IFN-γ. C57BL/6 mice were vaccinated with PPD, CFP-10–ESAT-6, and CFP-10–ESAT-6–PPE68′. The control group was immunized with saline. IFN-γ levels were measured in triplicate by ELISA. The results are expressed as means ± SD. The IFN-γ levels of the CFP-10–ESAT-6–PPE68′-immunized mouse group was high at >6,500 pg/ml between 3 and 5 weeks. ∗, P < 0.05 versus results for the control. This experiment was repeated twice with similar results.